R ile analize başlarken¹

1 R nerede kullanılır

• Veri düzenleme
• İstatistik analiz
• Web sayfası hazırlama (Statik/Dinamik)
• Sunum hazırlama
• Programlama
• Otomatik, periodik ve tekrarlanabilir rapor hazırlama
• pdf, html, ppt oluşturma
• tez yazma
• kitap yazma
• CV oluşturma
• poster hazırlama
• rapor şablonu oluşturma
• …

---

2 R generation

R yıllar içinde çok fazla değişim gösterdi

https://rss.onlinelibrary.wiley.com/doi/10.1111/j.1740-9713.2018.01169.x

:scale 30%

---

3 R yükleme

http://www.youtube.com/watch?v=XcBLEVknqvY

What is R?

---

3.1 R-project

https://cran.r-project.org/

---

3.2 RStudio

---

3.2.1 RStudio

https://www.rstudio.com/

https://www.rstudio.com/products/rstudio/download/

https://moderndive.com/2-getting-started.html

---

3.2.2 RStudio

---

3.2.3 RStudio eklentileri

• Discover and install useful RStudio addins

https://cran.r-project.org/web/packages/addinslist/README.html

https://rstudio.github.io/rstudioaddins/

devtools::install_github("rstudio/addinexamples", type = "source")

---

3.3 MacOS için

3.3.1 X11

https://www.xquartz.org/

3.3.2 Java OS

https://support.apple.com/kb/dl1572

---

4 R zor şeyler için kolay, kolay şeyler için zor

• R makes easy things hard, and hard things easy

• Aynı şeyi çok fazla şekilde yapmak mümkün

R Syntax Comparison::CHEAT SHEET

https://www.amelia.mn/Syntax-cheatsheet.pdf

---

#RStats — There are always several ways to do the same thing… nice example on with the identity matrix by @TeaStats https://t.co/O3GXdPiM32

— Colin Fay 🤘 (@_ColinFay) April 1, 2019

---

5 R paketleri

5.1 Neden paketler var

---

I love the #rstats community.
Someone is like, “oh hey peeps, I saw a big need for this mundane but difficult task that I infrequently do, so I created a package that will literally scrape the last bits of peanut butter out of the jar for you. It's called pbplyr.”
What a tribe.

— Frank Elavsky ᴰᵃᵗᵃ ᵂᶦᶻᵃʳᵈ (@Frankly_Data) July 3, 2018

---

---

5.2 Paketleri nereden bulabiliriz

• Available CRAN Packages By Name
  https://cran.r-project.org/web/packages/available_packages_by_name.html

• CRAN Task Views
  https://cran.r-project.org/web/views/

• Bioconductor
  https://www.bioconductor.org

• RecommendR
  http://recommendr.info/

• pkgsearch
  CRAN package search
  https://github.com/metacran/pkgsearch

• CRANsearcher
  https://github.com/RhoInc/CRANsearcher

• Awesome R
  https://awesome-r.com/

---

5.3 Kendi paket evrenini oluştur

• pkgverse: Build a Meta-Package Universe
  https://cran.r-project.org/web/packages/pkgverse/index.html

---

5.4 R paket yükleme

install.packages("tidyverse", dependencies = TRUE)
install.packages("jmv", dependencies = TRUE)
install.packages("questionr", dependencies = TRUE)
install.packages("Rcmdr", dependencies = TRUE)
install.packages("summarytools")

# install.packages("tidyverse", dependencies = TRUE)
# install.packages("jmv", dependencies = TRUE)
# install.packages("questionr", dependencies = TRUE)
# install.packages("Rcmdr", dependencies = TRUE)
# install.packages("summarytools")

---

5.5 Paket çağırma

# require(tidyverse)  
# require(jmv)  
# require(questionr)  
# library(summarytools)  
# library(gganimate)  

---

6 R için yardım bulma

# ?mean
# ??efetch
# help(merge)
# example(merge)

---

• Vignette

:scale 80%

---

• RDocumentation https://www.rdocumentation.org

• R Package Documentation https://rdrr.io/

• GitHub

• Stackoverflow

https://stackoverflow.com/

• Google uygun anahtar kelime

---

How I use #rstats
h/t @ThePracticalDev pic.twitter.com/erRnTG0Ujr

— Emily Bovee (@ebovee09) August 10, 2018

---

---

• Google’da ararken [R] yazmak da işe yarayabiliyor.

---

• searcher package 📦

---

• Awesome Cheatsheet https://github.com/detailyang/awesome-cheatsheet

http://cran.r-project.org/doc/contrib/Baggott-refcard-v2.pdf

https://www.rstudio.com/resources/cheatsheets/

• Awesome R

https://github.com/qinwf/awesome-R#readme

https://awesome-r.com/

• Twitter

https://twitter.com/hashtag/rstats?src=hash

---

• Use Reproducible Examples When Asking

Got a question to ask on @SlackHQ or post on @github? No time to read the long post on how to use reprex? Here is a 20-second gif for you to format your R codes nicely and for others to reproduce your problem. (An example from a talk given by @JennyBryan) #rstat pic.twitter.com/gpuGXpFIsX

— ZhiYang (@zhiiiyang) October 18, 2018

• Keeping up to date with R news
  https://masalmon.eu/2019/01/25/uptodate/

---

7 Rstudio ile proje oluşturma

https://support.rstudio.com/hc/en-us/articles/200526207-Using-Projects

---

8 RStudio ile veri yükleme

https://support.rstudio.com/hc/en-us/articles/218611977-Importing-Data-with-RStudio

---

8.1 Excel

8.2 SPSS

8.3 CSV

---

9 Veriyi görüntüleme

Spreadsheet users using #rstats: where's the data?#rstats users using spreadsheets: where's the code?

— Leonard Kiefer (@lenkiefer) July 7, 2018

---

10 Veriyi görüntüleme

# library(nycflights13)
# summary(flights)

View(data)

data

head

tail

glimpse

str

skimr::skim()

---

11 Veriyi değiştirme

11.1 Veriyi kod ile değiştirelim

11.2 Veriyi eklentilerle değiştirme

:scale 30%

---

11.3 RStudio aracılığıyla recode

questionr paketi kullanılacak

:scale 30%

---

https://juba.github.io/questionr/articles/recoding_addins.html

---

---

---

12 Basit tanımlayıcı istatistikler

summary()

mean

median

min

max

sd

table()

---

library(readr)
irisdata <- read_csv("data/iris.csv")

jmv::descriptives(
    data = irisdata,
    vars = "Sepal.Length",
    splitBy = "Species",
    freq = TRUE,
    hist = TRUE,
    dens = TRUE,
    bar = TRUE,
    box = TRUE,
    violin = TRUE,
    dot = TRUE,
    mode = TRUE,
    sum = TRUE,
    sd = TRUE,
    variance = TRUE,
    range = TRUE,
    se = TRUE,
    skew = TRUE,
    kurt = TRUE,
    quart = TRUE,
    pcEqGr = TRUE)

---

# install.packages("scatr")
scatr::scat(
    data = irisdata,
    x = "Sepal.Length",
    y = "Sepal.Width",
    group = "Species",
    marg = "dens",
    line = "linear",
    se = TRUE)

---

12.1 summarytools

https://cran.r-project.org/web/packages/summarytools/vignettes/Introduction.html

library(summarytools)
summarytools::freq(iris$Species, style = "rmarkdown")

---

summarytools::freq(iris$Species, report.nas = FALSE, style = "rmarkdown", headings = FALSE)

---

with(tobacco, print(ctable(smoker, diseased), method = 'render'))

---

with(tobacco,
     print(ctable(smoker, diseased, prop = 'n', totals = FALSE),
           omit.headings = TRUE, method = "render"))

---

summarytools::descr(iris, style = "rmarkdown")

---

descr(iris,
      stats = c("mean", "sd", "min", "med", "max"),
      transpose = TRUE,
      headings = FALSE,
      style = "rmarkdown")

---

# view(dfSummary(iris))

---

# First save the results
iris_stats_by_species <- by(data = iris,
                            INDICES = iris$Species,
                            FUN = descr, stats = c("mean", "sd", "min", "med", "max"),
                            transpose = TRUE)
# Then use view(), like so:
view(iris_stats_by_species, method = "pander", style = "rmarkdown")

---

# view(iris_stats_by_species)

---

data(tobacco) # tobacco is an example dataframe included in the package
BMI_by_age <- with(tobacco,
                   by(BMI, age.gr, descr,
                      stats = c("mean", "sd", "min", "med", "max")))
view(BMI_by_age, "pander", style = "rmarkdown")

---

BMI_by_age <- with(tobacco,
                   by(BMI, age.gr, descr,  transpose = TRUE,
                      stats = c("mean", "sd", "min", "med", "max")))

view(BMI_by_age, "pander", style = "rmarkdown", omit.headings = TRUE)

---

what.is(iris)

---

12.2 skimr

library(skimr)
skim(df)

---

12.3 DataExplorer

library(DataExplorer)
DataExplorer::create_report(df)

---

12.4 inspectdf

https://github.com/alastairrushworth/inspectdf

---

12.5 Grafikler

# library(ggplot2)
# library(mosaic)
# mPlot(irisdata)

descr(tobacco, style = 'rmarkdown')

print(descr(tobacco), method = 'render', table.classes = 'st-small')

dfSummary(tobacco, style = 'grid', plain.ascii = FALSE)

print(dfSummary(tobacco, graph.magnif = 0.75), method = 'render')

---

Here, building up a #ggplot2 as slowly as possible, #rstats. Incremental adjustments. #rstatsteachingideas pic.twitter.com/nUulQl8bPh

— Gina Reynolds (@EvaMaeRey) August 13, 2018

---

Dreaming of a fancy #Rstats #ggplot #dataviz but still scared of typing #code? @_pvictorr esquisse package has you covered https://t.co/1vIDXcVAAF pic.twitter.com/RlTkptnrNv

— Radoslaw Panczak (@RPanczak) October 2, 2018

---

13 Bazı arayüzler

https://sbalci.github.io/MyRCodesForDataAnalysis/R-Arayuzler.nb.html

---

13.1 Rcmdr

library(Rcmdr)

Rcmdr::Commander()

• A Comparative Review of the R Commander GUI for R

http://r4stats.com/articles/software-reviews/r-commander/

---

13.2 jamovi

https://www.jamovi.org/

https://blog.jamovi.org/2018/07/30/rj.html

---

14 R nereden öğrenilir

https://sbalci.github.io/MyRCodesForDataAnalysis/WhereToLearnR.nb.html

---

15 Sonraki Konular

• RStudio ile GitHub kullanımı
• R Markdown ve R Notebook ile tekrarlanabilir rapor
• Hipotez testleri

---

16 Geri Bildirim

• Geri bildirim için tıklayınız: Geri bildirim formu

---

---

17 Libraries Used

citation()

citation("tidyverse")
citation("foreign")
citation("tidylog")
citation("janitor")
citation("jmv")
citation("tangram")
citation("finalfit")
citation("summarytools")
citation("ggstatplot")
citation("readxl")

sessionInfo()

---

18 İletişim

Completed on 2019-09-24 22:55:33.

Serdar Balci, MD, Pathologist
drserdarbalci@gmail.com

https://rpubs.com/sbalci/CV
https://sbalci.github.io/
https://github.com/sbalci
https://twitter.com/serdarbalci

---

CommitMessage <- paste("updated on ", Sys.time(), sep = "")

wd <- getwd()

gitCommand <- paste("cd ", wd, " \n git add . \n git commit --message '", CommitMessage, "' \n git push origin master \n", sep = "")

system(command = gitCommand, intern = TRUE
)

---

1. Bu bir derlemedir, mümkün mertebe alıntılara referans vermeye çalıştım.↩︎